Women with endometriosis are at 50% increased risk of developing epithelial ovarian cancer, specifically endometrioid or clear-cell histotypes. Several studies suggest that concurrent endometriosis with ovarian cancer portrays prognostic benefit. The unique molecular mechanisms underlying the transformation of endometriosis to ovarian cancer is unknown but could be used to target novel therapies. Mouse and in vitro model systems that recapitulate the genetic make up of endometriosis-associated ovarian cancers are needed to understand the unique biology underlying these tumors. Loss-of-function mutations in ARID1A are common in endometriosis-associated ovarian cancers. However, loss of Arid1a alone in the mouse ovary is not sufficient to drive cancer. The rationale for adding oncogenic Kras to this mouse is 3 fold. One, KRAS signaling is significantly dysregulated in endometrioid ovarian cancers with concurrent endometriosis compared to endometrioid ovarian cancers without concurrent endometriosis. Two, KRAS itself is frequently mutated in endometrioid ovarian cancers with concurrent endometriosis. Three, human endometrioid endometrial cancers frequently have co-occurrence of ARID1A and KRAS mutations. The long-term goal is to make a preclinical mouse model of endometriosis-associated ovarian cancer that recapitulates the clinical course and genetic make up in humans. The overall objective of this application is to understand the unique tumorigenic mechanism of loss of ARID1A with oncogenic KRAS in vivo and in vitro in the context of ovarian cancer. The rationale for studying this is that understanding the molecular mechanism between loss of ARID1A and oncogenic KRAS may be the key to personalized therapy and improved prognosis for tumors with concurrent endometriosis. Our central hypothesis is that deletion of ARID1A on a background of oncogenic KRAS will affect tumor biology and lead to endometriosis-associated ovarian cancers. Guided by strong preliminary data, this hypothesis will be tested by pursuing two specific aims: 1) To generate and characterize a mouse model with conditional loss of Arid1a as a precursor lesion with oncogenic Kras; 2) To define the molecular mechanism between ARID1A and KRAS in vitro. Under the first aim, an Amhr2CreArid1af/fKrasLSL-G12D mouse model will be created and characterized by how well this model recapitulates human disease using molecular IHC and RNA-Seq. Under the second aim, CRISPR/Cas9 technology will be used to create genetically modified ovarian cancer cell lines that are ARID1A-/-KRASG12D. These lines will be characterized at the cellular and molecular level. This contribution is significant because these novel mouse and in vitro models may be tapped for discovery or testing of novel pharmacologic therapies for genetically similar tumors. The proposed research is innovative because this genetic combination is created through development similar to what likely happens with endometriosis. Bioinformatics analysis from this cross-species (mouse-human) comparison is innovative. Ultimately, such a model has the potential to lead novel therapeutic options.